CLIP: A Dataset for Extracting Action Items for Physicians from Hospital Discharge Notes
8772Authors
nan
Abstract
AbstractContinuity of care is crucial to ensuring positive health outcomes for patients discharged from an inpatient hospital setting, and improved information sharing can help. To share information, caregivers write discharge notes containing action items to share with patients and their future caregivers, but these action items are easily lost due to the lengthiness of the documents. In this work, we describe our creation of a dataset of clinical action items annotated over MIMIC-III, the largest publicly available dataset of real clinical notes. This dataset, which we call CLIP, is annotated by physicians and covers 718 documents representing 100K sentences. We describe the task of extracting the action items from these documents as multi-aspect extractive summarization, with each aspect representing a type of action to be taken. We evaluate several machine learning models on this task, and show that the best models exploit in-domain language model pre-training on 59K unannotated documents, and incorporate context from neighboring sentences. We also propose an approach to pre-training data selection that allows us to explore the trade-off between size and domain-specificity of pre-training datasets for this task.
Keywords
Full Content
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, pages 114–127 August 1–6, 2021. ©2021 Association for Computational Linguistics 114 Generalising Multilingual Concept-to-Text NLG with Language Agnostic Delexicalisation Giulio Zhou Huawei Noah’s Ark Lab London, UK giuliozhou@huawei.com Gerasimos Lampouras Huawei Noah’s Ark Lab London, UK gerasimos.lampouras@huawei.com Abstract Concept-to-text Natural Language Generation is the task of expressing an input meaning rep- resentation in natural language. Previous ap- proaches in this task have been able to gener- alise to rare or unseen instances by relying on a delexicalisation of the input. However, this of- ten requires that the input appears verbatim in the output text. This poses challenges in mul- tilingual settings, where the task expands to generate the output text in multiple languages given the same input. In this paper, we ex- plore the application of multilingual models in concept-to-text and propose Language Agnos- tic Delexicalisation, a novel delexicalisation method that uses multilingual pretrained em- beddings, and employs a character-level post- editing model to inflect words in their correct form during relexicalisation. Our experiments across five datasets and five languages show that multilingual models outperform monolin- gual models in concept-to-text and that our framework outperforms previous approaches, especially in low resource conditions. 1 Introduction Recently, neural approaches to language generation have become predominant in various tasks such as concept-to-text Natural Language Generation (NLG), Summarisation, and Machine Translation thanks to their ability to achieve state-of-the-art performance through end-to-end training (Duˇsek et al., 2018; Chandrasekaran et al., 2019; Barrault et al., 2019). Specifically in Machine Translation, deep learning models have proven easy to adapt to multilingual output (Johnson et al., 2017) and have been demonstated to successfully transfer knowl- edge between languages, benefiting both the low and high resource languages (Dabre et al., 2020). In the concept-to-text NLG task, the language generation model has to produce a text that is an Input MR: X1 X2 X3 X4 broadcastedby firstaired lastaired where: X1=bananaman|бананамен, X2=bbc|би_би_си, X3=1983_10_03, X4=1986_04_15 Gold Target References: bananaman first aired on the bbc on october 3rd, 1983 and broadcast its last episode on april 15th, 1986. бананамен впервые вышел в эфир на bbc 3 октября 1983 года, а его последний эпизод вышел 15 апреля 1986 года. Таблица 1: Delexicalisation and relexicalisation strategies on English and Russian examples from the WebNLG 2020 Challenge. Exact Delexicalisation: X1 first aired on the X2 on october 3rd, 1983 and broadcast its last episode on april 15th, 1986. X1 впервые вышел в эфир на bbc 3 октября 1983 года, а его последний эпизод вышел 15 апреля 1986 года. Language Agnostic Delexicalisation (LAD): X1 first aired on the X2 on october X3 and broadcast its last episode on X4. X1 впервые вышел в эфир на X2 X3 года, а его послед- ний эпизод вышел X4 года. Таблица 2: Delexicalisation and relexicalisation strategies on English and Russian examples from the WebNLG 2020 Challenge. Generated Output before Relexicalisation: (assuming training with LAD) X1 first aired on the …
Content truncated to 500 words
Labeled Text
AbstractContinuity of care is crucial to ensuring positive health outcomes for patients discharged from an inpatient hospital setting, and improved information sharing can help. To share information, caregivers write discharge notes containing action items to share with patients and their future caregivers, but these action items are easily lost due to the lengthiness of the documents. In this work, we describe our creation of a dataset of clinical action items annotated over MIMIC-III, the largest publicly available dataset of real clinical notes. This dataset, which we call CLIP, is annotated by physicians and covers 718 documents representing 100K sentences. We describe the task of extracting the action items from these documents as multi-aspect extractive summarization, with each aspect representing a type of action to be taken. We evaluate several machine learning models on this task, and show that the best models exploit in-domain language model pre-training on 59K unannotated documents, and incorporate context from neighboring sentences. We also propose an approach to pre-training data selection that allows us to explore the trade-off between size and domain-specificity of pre-training datasets for this task.
Publication Details
- Paper ID:
- 8772
- Year:
- 2021
- Publication:
- TIPSTER
- Publication Type:
- conf
- DOI:
- nan